定义与应用
-
IoU(Intersection over Union):gt bbox、pred bbox 交集的面积 / 二者并集的面积
-
NMS:当在图像中预测多个proposals、pred bboxes时,由于预测的结果间可能存在高冗余(即同一个目标可能被预测多个矩形框),因此可以过滤掉一些彼此间高重合度的结果;具体操作就是根据各个bbox的score降序排序,剔除与高score bbox有较高重合度的低score bbox,那么重合度的度量指标就是IoU;
-
mAP:得到检测算法的预测结果后,需要对pred bbox与gt bbox一起评估检测算法的性能,涉及到的评估指标为mAP,那么当一个pred bbox与gt bbox的重合度较高(如IoU score > 0.5),且分类结果也正确时,就可以认为是该pred bbox预测正确,这里也同样涉及到IoU的概念;
实现
- 计算 bbox 间 IoU 唯一的难点就在计算intersection,直接计算两个 bbox 的相交边框坐标即可,若不相交得到的结果中,必有 ixmax < ixmin 或者 iymax < iymin 其一成立,此时 iw、ih 就为 0 了
- 计算union,union = s1 + s2 - intersection
- 源码
# -*- coding: utf-8 -*-
# Author: zcj 2021.9.6
import numpy as np
def get_IoU(pred_bbox, gt_bbox):
"""
return iou score between pred / gt bboxes
:param pred_bbox: predict bbox coordinate
:param gt_bbox: ground truth bbox coordinate
:return iou score
"""
assert ((pred_bbox[2] - pred_bbox[0]) > 0 and
(pred_bbox[3] - pred_bbox[1]) > 0)
assert ((gt_bbox[2] - gt_bbox[0]) > 0 and
(gt_bbox[3] - gt_bbox[1]) > 0)
# -----0----- get coordinates of inters
ixmin = max(pred_bbox[0], gt_bbox[0])
iymin = max(pred_bbox[1], gt_bbox[1])
ixmax = min(pred_bbox[2], gt_bbox[2])
iymax = min(pred_bbox[3], gt_bbox[3])
iw = np.maximum(ixmax - ixmin + 1., 0.)
ih = np.maximum(iymax - iymin + 1., 0.)
# -----1----- intersection
inters = iw * ih
# -----2----- union, uni = s1 + s2 - inters
uni = ((pred_bbox[2] - pred_bbox[0] + 1.) * (pred_bbox[3] - pred_bbox[1] + 1.) +
(gt_bbox[2] - gt_bbox[0] + 1.) * (gt_bbox[3] - gt_bbox[1] + 1.) - inters)
# -----3----- iou
overlaps = inters / uni
return overlaps
def get_max_IoU(pred_bboxes, gt_bbox):
"""
given 1 gt bbox, >1 pred bboxes, return max iou score for the given gt bbox and pred_bboxes
:param pred_bbox: predict bboxes coordinates, we need to find the max iou score with gt bbox for these pred bboxes
:param gt_bbox: ground truth bbox coordinate
:return: max iou score
"""
if pred_bboxes.shape[0] > 0:
# -----0----- get coordinates of inters, but with multiple predict bboxes
ixmin = np.maximum(pred_bboxes[:, 0], gt_bbox[0])
iymin = np.maximum(pred_bboxes[:, 1], gt_bbox[1])
ixmax = np.minimum(pred_bboxes[:, 2], gt_bbox[2])
iymax = np.minimum(pred_bboxes[:, 3], gt_bbox[3])
iw = np.maximum(ixmax - ixmin + 1., 0.)
ih = np.maximum(iymax - iymin + 1., 0.)
# -----1----- intersection
inters = iw * ih
# -----2----- union, uni = s1 + s2 - inters
uni = ((pred_bboxes[:, 2] - pred_bboxes[:, 0] + 1.) * (pred_bboxes[:, 3] - pred_bboxes[:, 1] + 1.) +
(gt_bbox[2] - gt_bbox[0] + 1.) * (gt_bbox[3] - gt_bbox[1] + 1.) - inters)
# -----3-----iou, get max score and max iou index
overlaps = inters / uni
ovmax = np.max(overlaps)
jmax = np.argmax(overlaps)
return overlaps, ovmax, jmax
if __name__ == "__main__":
pass
# test1
pred_bbox = np.array([50, 50, 90, 100])
gt_bbox = np.array([70, 80, 120, 150])
print(get_IoU(pred_bbox, gt_bbox))
# test2
pred_bboxes = np.array([[15, 18, 47, 60],
[50, 50, 90, 100],
[70, 80, 120, 145],
[130, 160, 250, 280],
[25.6, 66.1, 113.3, 147.8]])
gt_bbox = np.array([70, 80, 120, 150])
print(get_max_IoU(pred_bboxes, gt_bbox))